DEB OHAH TANNEN

FR..ANES AND SCHEMAS IN INTERACTION
The terms and concepts of frame, script, and schema, as well as associated
ones such as prototype, speech activity [Gumperz 1982], and template, have
been variously used in linguistics (fat example, Chafe [1977], Fillmore
[1976]), a.rtificial intelligence (for example, Minsky [1975], Schank and
son (1978], Winograd [1977]), cognitive psychology (for example, Rumelhart
[1975]), anthropology (Bateson [1972], Frake [1977], Hymes [1974]), and,
perhaps the best known, sociology [Goffman 1974]. Elsewhere [Tannen 1979]
I have surveyed and discussed these terms and their use in these various disci-
plines, suggesting that they all reflect the notion of 'structures of expectations'
[Ross 1975].
In doing my initial work on frames, I was inspired and influenced in my
thinki.ng by a group I then belonged to which called itself The Story Group
and met regularly in the San Francisco Bay Area. I was concerned at that time
almost entirely with the concept of frames as used in the fields represented by
members of that group: psychology, linguistics, and artificial intelligence -
that is, knowledge structures in the mind which influence and account for the
comprehension (the primary interest of psychologists and artificial intel-
ligentsia) and production (the primary interest of linguists) of discourse.
AlthOl.lgh I included the anthropological/sociological concept of frame in my
review of cross-disciplinary treatment of the term [Tannen 1979], I did not
make :much use of it in analyzing my data. This omission - or, more gener-
ously, and perhaps more accurately, this focus - was also facilitated (as is so
often the case) by the nature of the data I was then analyzing: mostly monolo-
gic narratives 1, whereas the anthropological concept of frame applies to inter-
pretation: of meaning in interaction.
This 1,'Japer is based on a longer study which contatns sometvhat less theoretical discussion of
frames and schemas, and constderably more examples from the data: D, Tannen and C. Wallat, In..
teractive Frames and Knowledge Structure Schemas in Interaction: Examples from a Pediatric Ex-
amination, presented at tbe US-France Joint Seminar on Natural Language   Cadarache,
France, June 1982. All analysis 0/ the VIdeotapes has been done jointly with Cynthia Wallat. Theoret-
ical disc/,/..fSzon and application 0/ frames and schemas is mzne. The videotapes under analysts were
made at the Chtld Development Center 0/ the medical school 0/ GeorgetouJn UnIversity. We are
grateful to the CDC sta/t and especzally to the pediatrtcian, the mother, and the parent coordinator
assigned to the case, for giving us permtSslOn to use the vtdeotapes and for taking the tttne to discuss
them wah us durtng playback sessions, As I worked on revision 0/ the present paper, I was deepLy
dIstressed and saddened by the news I bad just received of the death 0/ Eroing Goffman. ;ls I wrote
and revtJed. I tJJas contlnually reminded 0/ him, as hts thtnking has been so central to mine - and
e1Jeryone's - on the topic of frames. It seems a hopelessly meager gesture tn the face of such a loss,
but I loant to dedtcate this paper to Erving Coffman.
1 Since the narratives were told to someone, they necessarily took the participation of the hear-
er into account and therefore were interacttve, and this is considered in my analysis, but nearly all
the talking was done by one speaker whereas the hearer merely answered direct questions, and
that rninilnally.
()UADERNI DI SEMANTICl\ / Vol. VI, no. 2, December 1985
326
In lny recent work, however, I have, together with Cynthia Wallat, who is a
social psychologist, been analyzing highly interactive data: videotapes of en..
  among a family, their handicapped child, and members of clinical
staff at the Child Development Center of Georgetown University's Department
of Pe(liatrics. Driven by the desire to account for verbal interaction in these
data, [ almost immediately confronted the need for the anthropological con-
cept of frames. In the present paper, then, I explore this interactive notion of
frame and show how it relates theoretically to and intertwines in interaction
with the knowledge structure notion which was the focus of my earlier work.
This knowledge structure notion of frame, variously named 'frame', 'script',
and 'schema', I shall now call 'schema' in order to distinguish it from interac-
tive fr.ames.
Interactive Frames
The anthropological!sociological" or interactive notion of FRAMES may be
understood as 'frames of interpretation', referring to the implicit or explicit
superordinate message about «what is going on here» [Goffman 1974], or
what is being DONE by talk [Gumperz 1982] without which no utterance (or
other l>ehavior) could be interpreted. Building on the work of Bateson [1972],
Goffman [1974] offers the most comprehensive discussion and demonstration
of frarnes, which he defines as «definitions of a situation.. . built up in
accordance with principles of organization which govern events ...» [Goffman
1974:10] 2.
To use Bateson's [1972:179] classic example, a monkey needs (and is able)
to whether behavior encountered in another monkey which is tYpically
associated with one activity - for example, a bite, which is associated with
'combat' - is intended in a given instance as combat or rather as play. 'Play'
and 'combat' are alternative FRAMES, or definitions, of what is going on, which
give meaning to the behavior, biting. People are confronted with similar inter-
pretive tasks in communication. In order to comprehend any utterance, a
listener (and a speaker) must know within which frame it is intended: for ex-
ample, is an utterance intended within the frame of joking? Is it fighting? Is it
an insult? Moreover, something intended as a joke but interpreted as an insult
can trigger a fight. An utterance could of course be both, and being aware of
this duality, a speaker can camouflage an insult as a joke.
A mistake in identification of frames can be funny or devastating. The re-
sults serious, for example, in the famous incident in which Orson Welles'
radio performance of H. G. Wells' «War of the Worlds» was mistaken by
listeners who tuned in too late to catch the frame identification as a radio play
to be a real newscast about invaders from outer space. Intentionally misleading
establisllment of frames underlies humor [Raskin 1984].
2 I-Ie does not, however, in his initial work focus analysis on talk per se; however, a later article
[Goffman 1981] does.
327
Even unintentionally mistaken frames frequently have amusing results. For
example, I was talking to a friend on the telephone; in the middle of our
friendly chat, he suddenly yelled, «YOU STOP THAT!» I knew from the way he
yelled this that the interjection was not addressed to me, even though I could
not see him. I quickly surmised that it was addressed to a dog, and I remarked
that he addressed the dog in something that sounded like a southern accent.
My friend explained that this was because the dog had learned to respond to
commands from its owner in that accent, and he further illustrated this fact by
telling me how he invites the dog to play: «I say, 'GO GIT THAT BALL!'» The
dog, however, still in earshot, could not understand the words 'I say' that
identified the frame, «ILLUSTRATING how I talk to the dog when I want to
play with him», but recognized the signal of the frame 'PLAYING FETCH' and
  running about the room looking for something to fetch. (He settled
upon a stuffed frog). This example demonstrates, as well, that people (and
dogs) identify frames of interaction by both linguistic signals - that is, the
words uttered - and paralinguistic signals - that is, HOW they are uttered, or
what Gumperz [1982] calls «contextualization cues»: intonation, pitch, loud--
ness, voice quality, rate of speech, and so on. In the preceding example, the
way my friend uttered «you stop that» signalled the frame 'disciplining a pet'
rather than 'chatting with a friend'. The interactive notion of frame, then, re-
fers to a superordinate definition of what is being engaged in HOW a speaker
nleans what slbe says.
Structure Schemas
The term SCHEMA, as I will be using it, refers to KNOWLEDGE STRUCTURES
itl the minds of participants in interaction - expectations based on prior ex..
perience about objects, events, and settings. Linguistic semanticists have been
illterested in this phenomenon, as they have observed that even the literal
meaning of an utterance can be understood only by reference to a pattern of
prior knowledge. To give the often quoted example from Wittgenstein, the
\vord hammer can have no meaning to someone who has never seen a hammer
IJlsed. Or, to borrow an example from Fillmore [1976], the difference between
t:he phrases 'on land' and 'on the ground' can be understood only by reference
to a sequence of actions associated respectively with sailing and flying.
Moreover, the only way anyone can understand any discourse is by filling in
unstated information which is known from prior experience in the world. This
I)henomenon became clear to researchers in artificial intelligence as soon as
they tried to get computers to understand even the simplest discourse. For in-
stance, in the now famous example of Schank and Abelson [1978], if you say
you went into a restaurant and make reference to «THE waitress», you are
assuming your audience is familiar with the expectation that restaurants have
,vaitresses or waiters. If you say the hamburger was undercooked and you left
a small tip, you are assuming familiarity with a complex network of expecta-
tions about the responsibilities of the waitress and the activities appropriate to
328
a customer. All of these expectations about what goes on in a restaurant, cal-
led by Schank and Abelson a 'restaurant script', amount to structures of ex-
pectations or knowledge structures associated with a particular setting, res-
taurallts. The realization that much of what is necessary to understand dis-
course - indeed, much of what is understood from discourse - is not actual-
ly verbalized in the discourse, became clear to researchers in linguistics as soon
as began to study natural conversation (for example Lakoff [1973] and
Brawll and Levinson [1978] on politeness phenomena), and to speech act
theorists who consequently hit upon the notion of indirect speech acts [Searle
19751
At an earlier stage of this study, I referred to the interactive notion of frame
as 'dynamic' and the knowledge structure notion of schema as 'static'. Indeed,
it is easy to conceptualize a restaurant script, for example,1 as a static set of
expectations about what goes on in a restaurant - a set which could be made
explicit and (for the purposes of artificial intelligence) told to a computer or
(for tble purposes of foreign language learning) told to a student. However, I
have now rejected that conceptualization, remembering that all types of frames
and scripts are dynamic, as Bartlett [1932], whose work underlies much of
present day schema theory, pointed out, and as others (for example, Frake
[1977]) have emphasized. That is, expectations about objects, people, settings,
ways to interact, and anything else in the world are not formed and stored
once and for all but rather are continually checked against current experience
and revised.
Interactive Frames and Knowledge Structure Schemas Interact
The dynamic processes of forming, revising, and applying expectations a-
bout the signalling of frames of interpretation as well as knowledge structure
schemas about people, objects, and situations, co-operate in interaction to
make communication possible. To illustrate again with the example of my tele-
phone conversation with my friend who interjected an order to a dog in our
friendly chat: I used my familiarity with his use of paralinguistic cues to sur-
mise that he had shifted suddenly from the frame 'chatting with a friend' (in
this case, me). But how did I know that he was disciplining a dog and not a
cat or g; child? I had to draw not only upon my ability to identify his    
guistic features as associated with 'disciplining a non-adult creature', but also
upon m:y k,nowledge that my friend was taking care of someone's dog - part
of my SCHEMA about this friend's immediate situation - as well as how this
friend 'V{ould be likely to address me, a dog, or a child - that is, a SCHEMA
for his habitual way of signalling interactive frames. Had my schema included
the information that my friend had a small child, was allergic to dogs, and was
itl the habit of addressing his child with mock-serious imperiousness, I might
have int,erpreted the same linguistic cues as signalling the related frame, 'disci-
plining a misbehaving child'. Thus, knowledge structure schemas refer to sets
of ex,pectations about people, objects, and situations, including expectations
329
about  which  linguistic  and  paralinguistic  features  are  to  be  used  to  signal  in-
teractive frames,  that is,  how  any  given  utterance is  intended.  Thus we  have,  to 
complicate  matters,  two  somewhat  different  but  clearly  related  senses  of  the 
knowledge  structure  sense  of schema:  one  a  set  of expectations  about  objects, 
people,  and  situations  (as  in the  restaurant  script  or  my  friend's  taking  care  of 
a  dog),  and  the  other  a  set  of expectations  about  how  language  is  used  in  in-
teraction,  how  interaction  works  (as  in how  my  friend  would  talk.  to  me  or  a 
d.og).  This  latter  sense  then  comes  rather  close  to  the  'principles  of organiza-
tion'  which  Goffman,  as  quoted  earlier,  postulates  as  the basis  upon which  (in-
frames  are  'built  up'.  In  this  sense,  knowledge  structure  schemas 
account  for  the  routinized  nature  of talk  in interaction 3 and  for  the  identifica-
tion  of interactive  frames.  In  order  to  clarify  and  investigate  these  hypotheses, 
d1e  remainder  of this  paper  illustrates  the  operation  of frames  and  schemas  in 
a  videotapes  interaction  in a  medical  setting:  the  examination  of a  child  by  a 
,. l)ediatrician  in  the  presence  of  the  mother. 
Joe Study
The  current analysis  is  part of an  ongoing  project  analyzing  videotapes  made 
at  the  Child  Development  Center,  a  facility  of  the  Georgetown  University 
Medical  School 4.  The  central  focus  of analysis  is  the  videotape  and  transcrip-
tion  of an  interaction  in which  a  pediatrician  examines  an  eight-year-old  cere-
  palsied  child  whom  we  call  Jady,  in her  mother's  presence.  In  addition, 
the  pediatrician  and  the  mother  viewed  the  tapes  with  us,  and  they  as  well  as 
other  members  of  the  professional  staff  discussed  with  us  their  recollections 
a.nd  interpretations. 
Interactive Frames in the Pediatric Examination
Early  analysis  of the  videotape  of the  pediatrician  examining  Jody  (Tannen 
Wallat  [1982],  [in press])  indicated  that  the  pediatrician  used  three  dis-
tinct  linguistic  registers  in  addressing  each  of three  audiences.  When talking  to 
the  mother,  she  spoke  in  an  unmarked  conversational  register.  In  addressing 
the  child,  she  used  a  teasing  register  characterized  by  exaggerated  shifts  in
IJitch  and  prosody,  and  drawn  out  vowel  sounds.  In  stark  contrast  to  this  in...
tonationally  exaggerated  register,  the  pediatrician  used  a  markedly  flat  intona-
tion  to give  a  running  account  of the  findings  of her examination,  addressed  to 
flO  present  party,  intended  for  the  video  audience  of  pediatric  residents.  We 
call  this  'reporting  register'.  It  is  clear  that  the  mother  perceived  this  reporting 
register  as  reflecting  the  doctor's  engagement  in  significant  conduct  of  busi-
3  Widdowson  [in  press]  calls  this  sense  of  schema  a  'procedural  schema'.  
4  Findings  of previous  analysis  are  reported  in Tannen  and Wallat  [1982]  and  [in press],  Wallat  
and  Tannen  [1982]. 
330 
ness,  for  the mother never  interrupts  to  ask  a  question when the doctor is  talk.. 
ing  in this  mode.  (The  notion  of 'reporting register'  accounts  for  a  similar  phe-
nomenon  described  by  Cicourel  [1975]  in an  analysis  of a  medical  interview). 
The  pediatrician  shifts  from  one  register  to  another,  sometimes  abrupdy, 
sometimes  gradually.  In  the  following  example,  she  shifts  from  entertaining 
Jody  to  reporting  findings  and  back  to  managing  Jody  in  a  prosodically 
market!  teasing  tone 5: 
Dr.:  TIlat's  my  light.  
Child:  /This  goes  up  there./  
Dr.:  It  goes  up  there.  That's  right.  Now  while  we're  examining  her  head  we're  feeling  
fox'  lymph  nodes  in her neck  or for  any  masses ... okay ... also  you  palpate the 
midline  for  thyroid,  for  goiter  if  there's  any.  Now  let  us  look  in your  mouth. 
Okay?  With  my  light  can  you  open  up  real  big?  .. Oh,  bigger ... Oh  bigger-
... Bigger. 
Thus  it  is  clear  that  the  pediatrician  is  required  to  shift  registers  in dealing 
with  audiences..  However  the  situation  is  even  more  complex  than  that. 
As  Goffman  .[1981:  128]  points  out,  «participants,  over  the  course  of  their 
speakirlg,  constantly  change  their  footing»,  FOOTING  being  «another  way  of 
talking  about  a  change  in  our  frame  for  events»,  defined  as  «a  change  in  the 
  we  take  up  to  ourselves  and  the  others  present  as  expressed  in the 
way  manage  the  production  or  reception  of  an  utterance».  Not  only  does 
the pecHatrician  have  to  talk  differendy  to  each of three  audiences:  the mother, 
the  chiJld,  and  the  video  audience,  but  she  must  deal  with  each  of these  audi-
ences  jjl  various  ways,  depending  upon  the  frame  she  is  operating  within. 
For  this  interaction,  two  often  conflicting  frames  are  evident:  an  examina-
tion  frBlme,  and  a  consultation  frame 6.  Activities  which  appear  the  same  on 
the  surface  can have very  different  meanings  and  consequences  for  the  partici...
pants  if they  are  understood  as  associated  with  different  frames.  For  example, 
when  t1le  pediatrician  asks  the  mother  for  information  relevant  to  the  child's 
condition,  she  is  still  focused  on the examination  of the  child,  following  aO  pre-
set    of focuses  of attention..  However,  when  the mother  asks  the  pediatri
cian  questions  of her  own  - for  example,  about  a  skin  eruption  behind  the 
child's  )tight  ear,  and  the  doctor  looks  behind  the  ear  in order  to  answer  the 
mother's  question - then the  doctor must  interrupt the examination,  and  shift 
5  In  brackets  ([)  = simultaneous  speech;  Iwordsl in slashes  are  not  certain. 
tion  mark  (?)  ;;:  rising  intonation,  not  grammatical. or  semantic  question.  Period  (.)  ;;:  falling 
tonation,  not  grammatical  sentence.  Comma  (,)  indicates  intonation  (<<more  to  come»). 
Three  dots  (...)  indicate  pause  of  roughly  half  second. 
6  These  are  certainly  not  the  only  frames  in  this  interaction,  but  they  are  the  most  prominent, 
and  the  ones  most  clearly  contrasting  in  salience  for  the  mother  and  doctor.  That  is,  it  seems  that 
the  examination  frame  is  primary for  the doctor,  whereas  the  consultation frame  is  primary for  the 
mother.  Another  frame,  a  video  frame,  reflects  the  fact  that  the  interaction  is  being  videotaped. 
This  with  tl:te  examination frame,  in  that  the  pediatrician  reports  her findings  to the video 
audience  of pediatric residents.  It  also  overlaps  with  a  frame  for  social  encounters,  as  the doctor is 
(in  in  this  setting  and  monitors  the  readiness  of the  video  crew  as  well  as  managing  the 
child,  telling  the  mother  where  to  sit,  and  so  on. 
331 
to a role associated with consultation with the mother. What on the surface
to be instances of the same activity - examining the child - are seen
as very different when they are understood as selVing different frames. In the
first case the doctor is adhering to a pre-set sequence of events in the ex-
aJnination, and in the second she is interrupting that sequence to focus on
something else, following which she will have to recover her place in what she
to at the outset as the 'standard pediatric evaluation' that she is there to
perform.
An excerpt from the data will illustrate that the demands associated with the
consultation frame can conflict with those of the examination frame, and that
t}lese frames and associated demands are seen in linguistic evidence. The
IIlother has recently learned that Jody has an arteriovenous malformation in
her brain, and she asks the doctor how dangerous this condition is. The doc-
tor responds in a way that balances the demands of several frames:
Ivlother: I often worry about the danger involved too.
Dr.: LYes.
cause she's well I mean like right now, ... uh ...
in her present condition.
Dr.: Lmhm
I've often wondered about how dangerous they they
are to her right now.
I)octor: We: 11 ... urn ... the only danger would be from
bleeding. . .. From them. If there was any
rupture, or anything like that which can happen
... urn ... that would be the danger. . .. F6r that.
Mo.: Lmhm-.J
But they're ... mm ... not going to be
something that will get worse as time goes on.
  Oh I see.
I)octor: But they're just there. Okay?
By using circumlocutions, 'buffer language', hesitations, conditional
structions (<<that would be the danger»), and a negative statement (<<they're not
going to get worse»), the doctor succeeds in reassuring the mother (to the
eloctor's own surprise when she later viewed the videotape), although in the
I,rocess she appears unsure of herself. In contrast, when reporting to the staff,
the doctor refers to the same condition with obvious concern, emphasizing the
dangers of «sudden death, intracranial hemorrhage». In this context, she is
totally fluent and appears very confident and assertive.
It seems clear, then, that in the context of the examination, the doctor was
caught in a conflict between the demand of the consultation frame that she
give the mother the information requested based on her medical know-
ledge, and b) deal with the emotional impact of that information on the
mother, and demand of the examination frame that she ,not take a lot of time
out from examining the child. The information that the child's life is in danger
is likely to upset the mother and therefore require significant counseling time.
(Notice that it is the admirable sensitivity of this doctor that makes her aware
of the needs of both frames. According to this mother, many doctors have in-
332
formed her in matter of fact tones of potentially devastating information about
her c.hild's condition, without showing any sign of awareness that such
formation will have emotional impact on the parent. In our terms, such doc-
tors acknowledge only one frame - examination - in order to avoid the de-
mands of conflicting frames - consultation and social encounter. Observing
the burden on this pediatrician, who does balance the demands of multiple
frames, makes it easy to understand why others might avoid this).
Knowledge Structure Schemas in the Pediatric Interaction
Knowledge structure mismatches account for extensive demands on the
pediatrician to switch from the examination to the consultation frame. For ex-
ample,,) when Jody sleeps, her breathing sounds noisy, as if she were gasping
for air. The mother is very concerned that the child might not be getting
enoug)1 oxygen, and when the doctor finishes examining the child's throat and
moves on to examine her ears, the mother takes the opportunity to interrupt
and introduce the consultation frame, by stating her concern. The doctor halts
the examination, turns to the mother, and explains that cerebral palsy entails
poor lnuscle control, and that this includes the muscles used in breathing;
therefore Jady's breathing sounds 'coarse' or 'floppy'. However, this does not
mean that she is having trouble breathing. The mother's schemas for health
and   palsy do not give her the expectation that the child's breathing
should sound noisy. In addition, she has a schema for 'wheezing' that leads
her to believe that noisy breathing is associated with difficulty breathing. In
fact, tIle parents, in the initial medical interview with a parent coordinator,
characterize lody as having difficulty l?reathing, and this is entered into the
written record of the intetview. These schemas, based on the parents' experi..
ence "lith non-cp children and limited experience with cp children, are not
easily altered. The pediatrician's assurance that Jody is not having trouble
breathing goes on for some time because of the mother's repetition of her con-
cern. Nonetheless, the mother brings it up agaih when the doctor is listening
to Jody's chest through a stethoscope, and again the doctor shifts from the
examination frame to a consultation frame to reassure her at length that the
child is not having trouble breathing, that these sounds are '.normal' for a child
with ce'rebral palsy. Yet again, when the pediatrician makes her report to the
parents.. the mother voices her concern that the child is having trouble breath-
ing and refers to the sound of Jody's breathing as 'wheezing'. At this point
the doctor adamantly reasserts that there is no wheezing. (What for the
mother is a general descriptive term for the sound of noisy breathing is for the
doctor a technical term denoting a condition by which the throat passages are
constricted). Thus the mismatched schemas for health and for cerebral palsy
continue to require frame shifts to consultation in contexts that are framed
differently for the professionals. Because of these differences in knowledge
structure schemas, the mother is easily reassured with respect to a condition
that is truly dangerous, the arteria-venous malformation, but cannot be reas-
33.3
suted with respect to a condition that is not, the noisy breathing. It is likely,
ironically, thatthevery reason the doctor employs exceedingly reassuring para-
lilJlguistic features in answering the .mother's questions abouth the a-v mal-
formation is that her schema informs her ofthe very real danger. Perhaps the
doctor is somewhat less reassuring (in terms ofparalinguistics, not in terms of
information given) in answering the mother's questions about the breathing
the doctor's schema gives her the assumption that condition is not
dangerous.
Summary and  Conclusion 
I have suggested that the terms FRAME and SCHEMA be used to refer to and
distinguish between an interactive and a knowledge structure notion ofstruc-
tU.res of expectation, respectively, and that an understanding of frames and
schemas helps account for interaction in a pediatric interview/examination.
Thereis every reason to believe that such frames and schemas operate in simi-
larways in all face toface interaction, although theparticularframes and sche-
m:,as will necessarily differ in different settings, and we may expect the usual
in.dividual and social differences both in the frames and in the paralinguistic
alld nonverbal features by which they are signalled.
REFERENCES
Bartlett, Frederic C. [1932], Remembering,  Cambridge, Cambridge University Press.
Bateson, Gregory [1972], Steps  to  an  ecology  of mind,  NY, Ballantine.
Brown, Penelope, and Stephen Levinson [1978], Universals  in  language  usage:  Polite-
ness  phenomena,  in Esther Goody (ed.), Questions  and politeness}  Cambridge, Cam-
bridge University Press, pp. 56-289.
Chafe, Wallace [1977], Creativity  in  verbalization  and  its  implications  for  the  nature  of 
stored knowledge,  in R. Freedle (ed.), Discourse  production  and comprehension,  Nor-
wood, N], Ablex, pp. 41-55.
CicoureI, Aron [1975], Discourse  and text:  Cognitive  and linguistic processes  in  studies  of 
social  structure,  in «Versus» 12, pp. 33-84.
Fillmore, Charles J. [1976], The  need for  a  frame  semantics  within  linguistics,  Stock..
holm, Skriptor, pp. 5-29.
Frake, Charles O. [1977], Plying frames  can  be  dangerous:  Some  reflections  on  methodol-
ogy  in  cognitive  anthropology,  in «The Quarterly Newsletter of the Institute for
Comparative Human Cognition», 1, pp. 1-7.
  Erving [1974], Frame  analysis,  NY, Harper & Row.
-- [1981], Footing.  Forms  0/ talk,  Philadelphia, University of Pennsylvania Press, pp.
124-159.
  JohnJ. [1982], Discourse  strategies,  Cambridge, Cambridge University Press.
lIymes, Dell [1974], Ways  0/ speaking,  in R. Bauman & J. Sherzer (eds.), Explorations 
in  the  ethnography  0/ speaking,  Cambridge, Cambridge University Press, pp. 433-
451.
Lakoff, Robin [1973], The  logic  0/ politeness

or  minding  your  pJs and q)sJ  Papers from
the ,Ninth Regional Meeting ofthe Chicago Linguistics Society, pp. 292-305.
I\linsky, Marvin [197.5], A  framework  for  representing  knowledge,  in P.H. Winston
(ed.), The  psychology  0/ computer  vt:fion,  New York, McGraw lIllI, pp.
Haskin, Victor [1984], Semantic  Mechanism.r  o/Humor,  Dordrecht - Boston - Lan-
caster, D. Reidel.
334
Ross, Robert N. [1975], Ellipsis and the structure 0/ expectation, San Jose State Occa-
sionf1u Papers in Linguistics, 1, pp. 183-191.
Rumelhart, David E. [1975], Notes on a schema for stories, in Daniel G. Bobrow &
Allan Collins (eds.), Representation and understanding, NY, Academic Press, pp.
211-236.
Schank, Roger C. & Robert P. Abelson [1978], Scripts, plans, goals, and understanding:
An tnquiry into human knowledge structures, Hillsdale, NJ, Erlbaum.
Searle;J John [1975], Indirect speech acts, in Peter Cole & Jerry Morgan (eds.), Syntax
and remanties, vol. 3: Speech Acts, NY, Academic, pp. 59-82.
Tannen, Deborah [1979], What)s in a frame? Sur/ace evidence for underlying expecta-
tions. in Roy Freedle (ed.), New directions in discourse processing, Norwood, NJ,
Ablex, pp. 137-181.
- & Cynthia Wallat [1982], A sociolinguistic analysis 0/ multiple demands on the
pediatrician in doctor/mother/patient interaction, in Robert Di Pietro (ed.), Linguis-
tics u ~   the professions, Norwood, NJ, Ablex, pp. 39-50.
- & Cynthia Wallat, [in press], Doctorllnotherlchtld communication: Linguistic analysis
0/ it pediatric interaction, in Sue Fisher & Alexandra Dundas Todd (eds.), The social
organization 0/ doctor/patient communication, Washington, D. C., The Center for
Applied Linguistics.
Wallat, C:ynthia & Deborah Tannen [1982], The dilemma 0/ parent part£cipation in
medical settings: A linguistic analysis, presented at American Sociological Associa-
tion Annual Meeting, San Francisco, September 1982.
Widdowson, Henry [in press], Procedures for discourse processing, in Chris Candlin
(ed.), Proceedings 0/ conference, Interpretive Strategies.
Winograd, Terry [1977], A Irarnework lor understanding discourse, in M. A. Just &
P. A. Carpenter (eds.), Cognitive processes in Comprehension, Hillsdale, NJ,
Erlbaum.
Tannen, Deborah. "Frames and Schemas in Interaction." "Quaderni di Semantica's Round
Table discussion on frame/script semantics," ed. by Victor Raskin. Quaderni di
Semantica 6(1985):2.326-335.

Sign up to vote on this title
UsefulNot useful